Ryan Wesslen
@ryanwesslen.bsky.social
ML Engineer. Data Scientist. HCI/Vis Researcher. Bayesian. Computational social scientist. Tar Heel. White Sox fan. Views are my own.
Reposted by Ryan Wesslen
Economics might seem -- from the outside -- like it's about competition. But really it's about creating the miracle of cooperation, where folks from all around the world enrich your day in a million tiny ways. It's that beauty that I'm worried about losing.
April 11, 2025 at 10:16 PM
Economics might seem -- from the outside -- like it's about competition. But really it's about creating the miracle of cooperation, where folks from all around the world enrich your day in a million tiny ways. It's that beauty that I'm worried about losing.
Reposted by Ryan Wesslen
behold the CONNECTED SPLATTERPIE
kneel before my works, ye mighty, and despair
kneel before my works, ye mighty, and despair
March 30, 2025 at 9:00 PM
behold the CONNECTED SPLATTERPIE
kneel before my works, ye mighty, and despair
kneel before my works, ye mighty, and despair
Reposted by Ryan Wesslen
“Responsible AI” is a bad word at NIST now.
Scientists that partner with the US Artificial Intelligence Safety Institute received new instructions that eliminate mention of “AI safety,” “responsible AI,” and “AI fairness” in the skills it expects of members @willknight.bsky.social www.wired.com/story/ai-saf...
Under Trump, AI Scientists Are Told to Remove ‘Ideological Bias’ From Powerful Models
A directive from the National Institute of Standards and Technology eliminates mention of “AI safety” and “AI fairness.”
www.wired.com
March 14, 2025 at 11:37 PM
“Responsible AI” is a bad word at NIST now.
Reposted by Ryan Wesslen
NEW: A searing editorial in the The Kyiv Independent.
“It’s time to say it plainly. America’s leadership has switched sides in the war. The American people have not, and they should speak up.
“A president just disrespected America in the Oval Office. It wasn’t Zelensky.”
@kyivindependent.com
“It’s time to say it plainly. America’s leadership has switched sides in the war. The American people have not, and they should speak up.
“A president just disrespected America in the Oval Office. It wasn’t Zelensky.”
@kyivindependent.com
Editorial: A president just disrespected America in the Oval Office. It wasn’t Zelensky
It’s time to say it plainly. America’s leadership has switched sides in the war. The American people have not, and they should speak up.
In the past several weeks, the U.S. leadership has demonstrate...
kyivindependent.com
February 28, 2025 at 10:45 PM
NEW: A searing editorial in the The Kyiv Independent.
“It’s time to say it plainly. America’s leadership has switched sides in the war. The American people have not, and they should speak up.
“A president just disrespected America in the Oval Office. It wasn’t Zelensky.”
@kyivindependent.com
“It’s time to say it plainly. America’s leadership has switched sides in the war. The American people have not, and they should speak up.
“A president just disrespected America in the Oval Office. It wasn’t Zelensky.”
@kyivindependent.com
Reposted by Ryan Wesslen
New paper <3
Interested in inference-time scaling? In-context Learning? Mech Interp?
LMs can solve novel in-context tasks, with sufficient examples (longer contexts). Why? Bc they dynamically form *in-context representations*!
1/N
Interested in inference-time scaling? In-context Learning? Mech Interp?
LMs can solve novel in-context tasks, with sufficient examples (longer contexts). Why? Bc they dynamically form *in-context representations*!
1/N
January 5, 2025 at 3:49 PM
New paper <3
Interested in inference-time scaling? In-context Learning? Mech Interp?
LMs can solve novel in-context tasks, with sufficient examples (longer contexts). Why? Bc they dynamically form *in-context representations*!
1/N
Interested in inference-time scaling? In-context Learning? Mech Interp?
LMs can solve novel in-context tasks, with sufficient examples (longer contexts). Why? Bc they dynamically form *in-context representations*!
1/N
Reposted by Ryan Wesslen
New paper in Political Analysis on synthetic text data for training classifiers. Main idea: generate training examples with LLMs, then fit classifiers on synthetic (+real) text. Paper has validations and guidance.
Blog: andrewhalterman.com/post/synthet...
Paper: www.cambridge.org/core/journal...
Blog: andrewhalterman.com/post/synthet...
Paper: www.cambridge.org/core/journal...
Synthetically generated text for supervised text analysis | Political Analysis | Cambridge Core
Synthetically generated text for supervised text analysis
www.cambridge.org
January 31, 2025 at 5:07 PM
New paper in Political Analysis on synthetic text data for training classifiers. Main idea: generate training examples with LLMs, then fit classifiers on synthetic (+real) text. Paper has validations and guidance.
Blog: andrewhalterman.com/post/synthet...
Paper: www.cambridge.org/core/journal...
Blog: andrewhalterman.com/post/synthet...
Paper: www.cambridge.org/core/journal...
Reposted by Ryan Wesslen
What are some of the things you've learned about how LLMs (and LLM-powered systems like ChatGPT) work that were non-obvious but most helped you build a more effective mental model of how to use them?
January 4, 2025 at 8:40 PM
What are some of the things you've learned about how LLMs (and LLM-powered systems like ChatGPT) work that were non-obvious but most helped you build a more effective mental model of how to use them?
Reposted by Ryan Wesslen
Wrote down the process to build your own custom feeds for Bluesky programmatically in Python and run it 100% free
Uses @skyfeed.app + @github.com actions to do periodic filtering and re-ranking and @cloudflare.social static pages to provide data to @bsky.app
Uses @skyfeed.app + @github.com actions to do periodic filtering and re-ranking and @cloudflare.social static pages to provide data to @bsky.app
Zero-Cost Custom Feeds on Bluesky
A simple stack for generating custom feeds for Bluesky programmatically without a backend server
amitness.com
December 1, 2024 at 2:43 PM
Wrote down the process to build your own custom feeds for Bluesky programmatically in Python and run it 100% free
Uses @skyfeed.app + @github.com actions to do periodic filtering and re-ranking and @cloudflare.social static pages to provide data to @bsky.app
Uses @skyfeed.app + @github.com actions to do periodic filtering and re-ranking and @cloudflare.social static pages to provide data to @bsky.app
Reposted by Ryan Wesslen
There’s something fundamentally wacky going on here. That a) there’s an arrow of time in language is cool but b) that the magnitude varies a lot between languages? Wacky.
February 12, 2024 at 10:02 PM
There’s something fundamentally wacky going on here. That a) there’s an arrow of time in language is cool but b) that the magnitude varies a lot between languages? Wacky.
Reposted by Ryan Wesslen
New post: Have been meaning to write something around what has fundamentally changed around the process of putting ML into prod now that we have LLMs. TL;DR: It's still just compression, we just don't control as much anymore.
vickiboykis.com/2024/01/15/w...
vickiboykis.com/2024/01/15/w...
What's new with ML in production
What's different about LLMs versus traditional ML
vickiboykis.com
January 18, 2024 at 8:45 PM
New post: Have been meaning to write something around what has fundamentally changed around the process of putting ML into prod now that we have LLMs. TL;DR: It's still just compression, we just don't control as much anymore.
vickiboykis.com/2024/01/15/w...
vickiboykis.com/2024/01/15/w...
Reposted by Ryan Wesslen
In 2023, we rolled out Prodigy v1.12-1.14 packed with new features like spacy-llm integration, prompt engineering, QA support like IAA metrics, task routing, and new plugins such as PDF and Hugging Face 🤗.
We highlight the many updates in our new blog post 🎉
explosion.ai/blog/prodigy...
We highlight the many updates in our new blog post 🎉
explosion.ai/blog/prodigy...
Prodigy in 2023: LLMs, task routers, QA and plugins · Explosion
We have made a ton of new updates in Prodigy this year with v1.12, v1.13, and v1.14 releases. So we decided to write a post about them.
explosion.ai
November 29, 2023 at 3:34 PM
In 2023, we rolled out Prodigy v1.12-1.14 packed with new features like spacy-llm integration, prompt engineering, QA support like IAA metrics, task routing, and new plugins such as PDF and Hugging Face 🤗.
We highlight the many updates in our new blog post 🎉
explosion.ai/blog/prodigy...
We highlight the many updates in our new blog post 🎉
explosion.ai/blog/prodigy...
Reposted by Ryan Wesslen
The PyData NYC video from "Half hour of labeling power: Can we beat GPT?" by @ryanwesslen.bsky.social & me is now live!
We show how to use LLMs to speed up annotation, collect 1.2k examples & beat our baseline.
📺 Video: www.youtube.com/watch?v=Ta45...
📝 Slides: speakerdeck.com/inesmontani/...
We show how to use LLMs to speed up annotation, collect 1.2k examples & beat our baseline.
📺 Video: www.youtube.com/watch?v=Ta45...
📝 Slides: speakerdeck.com/inesmontani/...
November 27, 2023 at 12:09 PM
The PyData NYC video from "Half hour of labeling power: Can we beat GPT?" by @ryanwesslen.bsky.social & me is now live!
We show how to use LLMs to speed up annotation, collect 1.2k examples & beat our baseline.
📺 Video: www.youtube.com/watch?v=Ta45...
📝 Slides: speakerdeck.com/inesmontani/...
We show how to use LLMs to speed up annotation, collect 1.2k examples & beat our baseline.
📺 Video: www.youtube.com/watch?v=Ta45...
📝 Slides: speakerdeck.com/inesmontani/...
Deck the halls with NLP! @explosion-ai.bsky.social's💀 t-shirts, tot bags, or mugs are sure to bring a touch of skeleton swagger this holiday season
Whenever we're at conferences, people ask me where they can buy our swag to support us. So we FINALLY DID IT!
Launching our new merch store, including:
📚 2 exclusive designed books
👕 supporter shirts
☕ mug, keychains, poster
💙 & of course stickers!
explosion.ai/merch
Launching our new merch store, including:
📚 2 exclusive designed books
👕 supporter shirts
☕ mug, keychains, poster
💙 & of course stickers!
explosion.ai/merch
November 20, 2023 at 1:47 PM
Deck the halls with NLP! @explosion-ai.bsky.social's💀 t-shirts, tot bags, or mugs are sure to bring a touch of skeleton swagger this holiday season
Reposted by Ryan Wesslen
Hello NYC ✨ Looking forward to seeing everyone at PyData NYC next week!
We'll have an @explosion-ai.bsky.social booth again with brand new swag, and @ryanwesslen.bsky.social & I will start the conf with our tutorial "Half hour of labeling power: Can we beat GPT?": nyc2023.pydata.org/cfp/talk/WQY...
We'll have an @explosion-ai.bsky.social booth again with brand new swag, and @ryanwesslen.bsky.social & I will start the conf with our tutorial "Half hour of labeling power: Can we beat GPT?": nyc2023.pydata.org/cfp/talk/WQY...
October 29, 2023 at 2:50 PM
Hello NYC ✨ Looking forward to seeing everyone at PyData NYC next week!
We'll have an @explosion-ai.bsky.social booth again with brand new swag, and @ryanwesslen.bsky.social & I will start the conf with our tutorial "Half hour of labeling power: Can we beat GPT?": nyc2023.pydata.org/cfp/talk/WQY...
We'll have an @explosion-ai.bsky.social booth again with brand new swag, and @ryanwesslen.bsky.social & I will start the conf with our tutorial "Half hour of labeling power: Can we beat GPT?": nyc2023.pydata.org/cfp/talk/WQY...
Reposted by Ryan Wesslen
Announcing ✨Prodigy-HF ✨
It's a new plugin that allows you to train @huggingface.bsky.social NER models directly on annotated data in Prodigy. It also provides a recipe to upload annotations to Hugging Face HUB!
It's a new plugin that allows you to train @huggingface.bsky.social NER models directly on annotated data in Prodigy. It also provides a recipe to upload annotations to Hugging Face HUB!
October 25, 2023 at 1:52 PM
Announcing ✨Prodigy-HF ✨
It's a new plugin that allows you to train @huggingface.bsky.social NER models directly on annotated data in Prodigy. It also provides a recipe to upload annotations to Hugging Face HUB!
It's a new plugin that allows you to train @huggingface.bsky.social NER models directly on annotated data in Prodigy. It also provides a recipe to upload annotations to Hugging Face HUB!
Reposted by Ryan Wesslen
Interesting attempt to categorize LLM hallucinations (from an international team including Stanford and Amazon) arxiv.org/abs/2310.04988
October 24, 2023 at 12:20 PM
Interesting attempt to categorize LLM hallucinations (from an international team including Stanford and Amazon) arxiv.org/abs/2310.04988
Reposted by Ryan Wesslen
I added support for PDFs for Prodi.gy in the past few weeks. So I figured I'd record a small demo.
www.youtube.com/watch?v=rwyz...
If folks are interested in working on detection models for PDFs -> let me know if there's recipes missing!
www.youtube.com/watch?v=rwyz...
If folks are interested in working on detection models for PDFs -> let me know if there's recipes missing!
Prodigy-PDF for PDF annotation and OCR - Prodigy Shorts
We've recently introduced Prodigy Plugins which extend the features of Prodigy by adding direct support for 3rd party integrations. One of these plugins is P...
www.youtube.com
October 24, 2023 at 3:29 PM
I added support for PDFs for Prodi.gy in the past few weeks. So I figured I'd record a small demo.
www.youtube.com/watch?v=rwyz...
If folks are interested in working on detection models for PDFs -> let me know if there's recipes missing!
www.youtube.com/watch?v=rwyz...
If folks are interested in working on detection models for PDFs -> let me know if there's recipes missing!
Reposted by Ryan Wesslen
We recently released ✨ Prodigy-ANN ✨ that allows you to use contextual search to find relevant subsets of data to annotate first.
To help explain this new feature, @koaning.bsky.social made a small demo to highlight the new feature 👀
youtu.be/jyu2nbjwfXw
To help explain this new feature, @koaning.bsky.social made a small demo to highlight the new feature 👀
youtu.be/jyu2nbjwfXw
October 20, 2023 at 1:56 PM
We recently released ✨ Prodigy-ANN ✨ that allows you to use contextual search to find relevant subsets of data to annotate first.
To help explain this new feature, @koaning.bsky.social made a small demo to highlight the new feature 👀
youtu.be/jyu2nbjwfXw
To help explain this new feature, @koaning.bsky.social made a small demo to highlight the new feature 👀
youtu.be/jyu2nbjwfXw
Reposted by Ryan Wesslen
The new OCR feature uses Pytesseract under the hood to attach parsed text to segments that you've annotated with the `pdf.image.manual` recipe.
If you want to learn more, the docs have plenty of extra info: 👀
prodi.gy/docs/plugins...
If you want to learn more, the docs have plenty of extra info: 👀
prodi.gy/docs/plugins...
October 19, 2023 at 12:47 PM
The new OCR feature uses Pytesseract under the hood to attach parsed text to segments that you've annotated with the `pdf.image.manual` recipe.
If you want to learn more, the docs have plenty of extra info: 👀
prodi.gy/docs/plugins...
If you want to learn more, the docs have plenty of extra info: 👀
prodi.gy/docs/plugins...
Reposted by Ryan Wesslen
Curious if your annotators are on the same page? Prodigy has just released v1.14.3 with built-in inter-annotator agreement (IAA) metrics to track and measure their agreement. In this 🧵, we'll review Prodigy's document-level IAA metrics. prodi.gy/docs/metrics
October 18, 2023 at 2:34 PM
Curious if your annotators are on the same page? Prodigy has just released v1.14.3 with built-in inter-annotator agreement (IAA) metrics to track and measure their agreement. In this 🧵, we'll review Prodigy's document-level IAA metrics. prodi.gy/docs/metrics